Find the mean for a group by summing all values of that group and dividing by the number of values.
Subtract the mean from each value of the group. Call this new set of data A.
Square each value in the dataset A.
Sum all of the squared values. This is your sum of squares for that group.
Repeat this process for all remaining groups.
Count the number of values for each group.
Sum all of these values. Call this number N.
Subtract the number of groups from N. This is your within groups degrees of freedom. Call this value DF.
Sum all of the sums of squares you calculated previously. Call this value SS.
Divide SS by DF. This is your pooled estimated variation.
Check your answer by taking its square root. This square root should be of similar size to your sample's individual standard deviations.